| { | |
| "best_global_step": 800, | |
| "best_metric": 0.10116679221391678, | |
| "best_model_checkpoint": "/tmp/results/checkpoint-800", | |
| "epoch": 0.8812999173781327, | |
| "eval_steps": 200, | |
| "global_step": 800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.055081244836133296, | |
| "grad_norm": 0.15732981264591217, | |
| "learning_rate": 0.00019640234948604994, | |
| "loss": 1.2703, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.11016248967226659, | |
| "grad_norm": 0.13457506895065308, | |
| "learning_rate": 0.00019273127753303965, | |
| "loss": 0.2101, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16524373450839988, | |
| "grad_norm": 0.11798923462629318, | |
| "learning_rate": 0.00018906020558002938, | |
| "loss": 0.1801, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.22032497934453318, | |
| "grad_norm": 0.10562872886657715, | |
| "learning_rate": 0.00018538913362701911, | |
| "loss": 0.1621, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22032497934453318, | |
| "eval_loss": 0.15831586718559265, | |
| "eval_runtime": 95.9669, | |
| "eval_samples_per_second": 33.626, | |
| "eval_steps_per_second": 4.21, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.27540622418066646, | |
| "grad_norm": 0.11607076972723007, | |
| "learning_rate": 0.00018171806167400882, | |
| "loss": 0.1526, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.33048746901679976, | |
| "grad_norm": 0.20103491842746735, | |
| "learning_rate": 0.00017804698972099853, | |
| "loss": 0.1498, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.38556871385293306, | |
| "grad_norm": 0.20092593133449554, | |
| "learning_rate": 0.00017437591776798826, | |
| "loss": 0.1481, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.44064995868906637, | |
| "grad_norm": 0.15526321530342102, | |
| "learning_rate": 0.000170704845814978, | |
| "loss": 0.1533, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.44064995868906637, | |
| "eval_loss": 0.13016295433044434, | |
| "eval_runtime": 96.1932, | |
| "eval_samples_per_second": 33.547, | |
| "eval_steps_per_second": 4.2, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.49573120352519967, | |
| "grad_norm": 0.23773688077926636, | |
| "learning_rate": 0.0001670337738619677, | |
| "loss": 0.1434, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5508124483613329, | |
| "grad_norm": 0.14756543934345245, | |
| "learning_rate": 0.00016336270190895743, | |
| "loss": 0.1316, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6058936931974662, | |
| "grad_norm": 0.19601435959339142, | |
| "learning_rate": 0.00015969162995594716, | |
| "loss": 0.1301, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.6609749380335995, | |
| "grad_norm": 0.1771874874830246, | |
| "learning_rate": 0.00015602055800293687, | |
| "loss": 0.1215, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6609749380335995, | |
| "eval_loss": 0.11715047806501389, | |
| "eval_runtime": 95.8683, | |
| "eval_samples_per_second": 33.661, | |
| "eval_steps_per_second": 4.214, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.7160561828697328, | |
| "grad_norm": 0.16557306051254272, | |
| "learning_rate": 0.00015234948604992657, | |
| "loss": 0.1139, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.7711374277058661, | |
| "grad_norm": 0.2927381992340088, | |
| "learning_rate": 0.0001486784140969163, | |
| "loss": 0.1214, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.8262186725419994, | |
| "grad_norm": 0.3338780701160431, | |
| "learning_rate": 0.00014500734214390604, | |
| "loss": 0.1285, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.8812999173781327, | |
| "grad_norm": 0.2900579273700714, | |
| "learning_rate": 0.00014133627019089574, | |
| "loss": 0.1274, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.8812999173781327, | |
| "eval_loss": 0.10116679221391678, | |
| "eval_runtime": 94.9307, | |
| "eval_samples_per_second": 33.993, | |
| "eval_steps_per_second": 4.256, | |
| "step": 800 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 2724, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 164707172352000.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |